Surgical Optimization in Preoperatively Low-risk cN1a PTC: A Predictive Model for High-Volume Central Lymph Node Metastasis.

BACKGROUND

Accurate preoperative identification of high-volume central lymph node metastasis (hv-CLNM; defined as more than 5 central lymph node metastases) is critical for guiding surgical decisions-lobectomy or total thyroidectomy-in patients with papillary thyroid carcinoma (PTC) clinically diagnosed with central neck lymph node metastasis (cN1a). Total thyroidectomy is generally preferred for patients with hv-CLNM. In contrast, lobectomy may be sufficient for patients with low-volume metastasis (5 or fewer lymph node metastases). This study aimed to identify predictors of hv-CLNM in preoperatively low-risk cN1a and to develop a predictive model to estimate the risk of hv-CLNM, thereby optimizing surgical decision-making.

METHODS

A total of 707 patients with pathologically confirmed PTC and classified as preoperatively low-risk cN1a were retrospectively enrolled. Clinical and ultrasound features were collected. Variables were selected using least absolute shrinkage and selection operator regression, followed by multivariate logistic regression to construct a predictive model. Internal validation was performed. Recurrence-free survival was compared between lobectomy and total thyroidectomy groups using propensity score matching.

RESULTS

Hv-CLNM occurred in 13.4% (96/707) of patients. Independent predictors of hv-CLNM included age, sex, tumor size, tumor location, and lymph node calcification. The nomogram demonstrated good discrimination (area under the plasma concentration-time curve = 0.75) and calibration. After adjustment, recurrence-free survival did not significantly differ between surgical groups.

CONCLUSIONS

This nomogram, based on readily available clinical and ultrasound features, effectively predicts the risk of hv-CLNM in preoperatively low-risk cN1a PTC. This tool may facilitate individualized surgical planning. Lobectomy appears to be a safe and appropriate option for most patients in this subgroup.